JOURNAL OF CLINICAL MICROBIOLOGY, Apr. 1993, p. 882-8860095-1137/93/040882-05$02.00/0Copyright © 1993, American Society for Microbiology

Detection of Hepatitis C Virus RNA by a Combined ReverseTranscription-Polymerase Chain Reaction AssayKAREN K. Y. YOUNG,`* ROBERT M. RESNICK,1 AND THOMAS W. MYERS2

Departments of Infectious Diseases' and Core Technology,2 Roche Molecular Systems, Inc.,1145 Atlantic Avenue, Alameda, California 94501

Received 23 September 1992/Accepted 19 January 1993

Amplification of RNA by the polymerase chain reaction (PCR) is normally a two-step process requiringseparate enzymes and buffer conditions. We describe a combined reverse transcription-PCR (RT-PCR) assayfor hepatitis C virus (HCV) RNA amplification in which a single enzyme and buffer condition are used. In thisassay, both the RT and PCR steps are carried out with the thermoactive DNA polymerase of hermnusthermophilus. A transcription vector containing HCV sequences has also been constructed to generatequantifiable HCV RNA templates that can be used to optimize reaction conditions and to assess the efficiencyof amplification. Amplification from 100 copies ofRNA was detected reproducibly by gel electrophoresis. Theassay sensitivity was increased to 10 RNA copies by hybridization to a probe. The patterns of viremia in threeindividuals infected with HCV were examined by amplification of HCV RNA from plasma samples collectedserially over a period of 1 year. These results were correlated with the times of seroconversion and the onsetof rise in levels of alanine aminotransferase in serum. In all three subjects, HCV RNA was detected prior toseroconversion and the initial rise in levels of alanine aminotransferase in serum. Upon seroconversion, HCVRNA fell to a level below the detection limit of the assay. This pattern of transient viremia appears to becharacteristic of acute, resolving HCV infections. The combined RT-PCR assay is a sensitive method whichcircumvents the problems associated with PCR amplification of RNA. Using this assay, we demonstrated thatthree donors infected by the same index case all have similar patterns of viremia.

Hepatitis C virus (HCV) is the major causative agent forparenterally transmitted non-A, non-B hepatitis (5). Detec-tion of HCV infection has been facilitated by the develop-ment of antibody detection assays. However, antibody de-tection methods are of restricted use because there is a meanwindow period of 22 weeks between infection and serocon-version (1). Loss of antibody in some persistently infectedindividuals has also been documented (4). Furthermore, dataon antibody status do not provide a means to differentiatebetween current, active infection; chronic infection; andpast, resolved infection. Methods which permit the directdetection of virus can augment information from antibodytesting. Amplification of viral nucleic acids by the poly-merase chain reaction (PCR) has been shown to be aneffective means for the direct detection of HCV (8, 18, 23).The genome of HCV is composed of a single-stranded

RNA molecule (5). As such, amplification by PCR must bepreceded by a step to generate a cDNA copy. The synthesisof the cDNA copy generally employs a retroviral reversetranscriptase. Amplification of the cDNA is typically per-formed by a thermostable and thermoactive DNA poly-merase from Thennus aquaticus. Previously, the require-ment for different enzymes in the two steps necessitated theaddition of new enzyme and change of buffer conditionsbetween the reverse transcription (RT) and amplificationsteps. This is cumbersome and increases the likelihood ofcontamination.We report here the development of an HCV RNA ampli-

fication assay in which both the RT and the amplificationsteps are carried out with one enzyme under a single set ofbuffer conditions. This combined RT-PCR assay takes ad-vantage of the fact that the DNA polymerase of the thermo-

* Corresponding author.

882

philic bacterium Thennus thennophilus possesses enhancedreverse transcriptase activity in the presence of manganese(17). Although optimal conditions for DNA and RNA tem-plates are different (17), a set of buffer conditions whichsupports both RT and PCR has been determined. The use ofa single enzyme and buffer condition obviates the need toopen reaction tubes between the RT and DNA amplificationsteps and thus minimizes the possibility of contamination.Coupled RT-PCR systems in which retroviral reverse tran-scriptase-mediated cDNA synthesis and Taq polymerase-mediated amplification are carried out in the presence ofboth enzymes under one buffer condition have been de-scribed (6, 10). However, the lower temperature necessaryfor retroviral reverse transcriptase activity may reduce theefficiency with which cDNA can be synthesized throughstable secondary RNA structures (2, 14). The elevatedreaction temperature required for the thermoactive T. ther-mophilus DNA polymerase used in the current assay woulddestabilize some of the secondary structure present in theRNA template and in addition increase the specificity ofprimer extension.The development of assays for the amplification of HCV

RNA has been hampered by the lack of well-defined tem-plates to optimize amplification parameters and to determinethe analytical sensitivity of the assay. To generate suchdefined templates, we have cloned the relevant portion of theHCV genome into a transcription vector. The concentrationof the HCV RNA transcribed in vitro was measured, andknown amounts were used as templates in amplificationreactions. This is the first description of an HCV RNAamplification assay in which the amplification conditions andanalytical sensitivity of the assay are completely character-ized.The combined RT-PCR assay was used to document the

natural history of HCV infection in three individuals who

Vol. 31, No. 4

on August 10, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

HEPATITIS C VIRUS RNA DETECTION BY COMBINED RT-PCR 883

were infected by contaminated blood products from thesame index case. Plasma samples collected serially at differ-ent time points postinfection were examined to determinethe pattern of viremia over time in each individual.

MATERIALS AND METHODS

Oligonucleotide primers and probe. The combined RT-PCR was performed with a primer pair selected from thehighly conserved 5' terminus of the HCV genome (11, 12,19). The pair consisted of upstream primer KY80, 5'-GCAGAAAGCGTCTAGCCATGGCGT (nucleotides [nt] 56 to 79[12]), and downstream primer KY78, 5'-CTCGCAAGCACCCTATCAGGCAGT (nt 276 to 299). (KY78 contains a biotinresidue at the 5' terminus to facilitate nonradioactive detec-tion of amplification products.) KY88, 5'-GTTGGGTCGCGAAAGGCCTTGTGGT (nt 251 to 275), served as the hybrid-ization probe.RNA template production. To construct an HCV RNA

transcription vector, HCV sequences from nt 50 to 599 (12)were cloned into the pSP64(poly A) vector (Promega, Mad-ison, Wis.). The identity of the final construct (pHCV1.1A)was confirmed by DNA sequence analysis.HCV template RNA was transcribed in vitro from

pHCV1.1A according to the manufacturer's instructions.Briefly, 5 ,ug of plasmid DNA was linearized and thenincubated with 40 U of SP6 RNA polymerase for 90 minat 37°C in the presence of 500 p.M (each) ribonucleosidetriphosphates (GTP, ATP, UTP, and CTP), 100 U ofRNasin, 10 mM dithiothreitol, 40 mM Tris-HCl (pH 7.5), 6mM MgCl2, 2 mM spermidine, and 10 mM NaCl in a totalreaction volume of 100 p.l. After the transcription reaction,the DNA template was degraded by two rounds of digestionwith RNase-free DNase (Promega) for 15 min at 37°C with 5and 10 U of enzyme, respectively. The RNA was dilutedwith 100 p.l of 10 mM Tris-HCl (pH 7.5)-i mM EDTA andadjusted to 240 mM sodium acetate (pH 4.0) to facilitatepartitioning of residual DNA into the organic phase duringsubsequent extraction with phenol (13). The HCV RNAtranscripts, which contained a poly(A) tail, were furtherpurified on an oligo(dT)-cellulose column (21). The RNAconcentration was determined spectrophotometrically byUV A260 (22). An aliquot was analyzed by agarose gelelectrophoresis to assess its integrity. A typical yield wasbetween 10 and 15 p,g of HCV RNA transcript. Serialdilutions of the RNA templates were made in the presence ofEscherichia coli rRNA (Sigma, St. Louis, Mo.) as the carrierat 10 ng/p,l.Combined RT-PCR and detection of amplification products.

The combined RT-PCR amplifications were carried out with20-,ul reaction mixtures containing 2 p.1 of template, 200 p.M(each) deoxyribonucleoside triphosphates (dNTPs) (dATP,dCTP, dGTP, and dTT'P), 0.85 mM MnCl2, 150 nM (each)primers KY78 and KY80, 5 U of T. thernophilus DNApolymerase (Perkin-Elmer, Norwalk, Conn.), and lx RT-PCR buffer (10x RT-PCR buffer consists of 100 mM Tris-HCl [pH 8.3] and 900 mM KCl). The reactions were per-formed in a GeneAmp PCR system 9600 thermocyclerutilizing thin-walled MicroAmp reaction tubes (Perkin-Elmer) without a mineral oil overlay. The thermocycler waspreheated to 70°C. RT was allowed to proceed for 15 min at70°C and was followed by a 1-min incubation at 95°C tofacilitate denaturation of RNA-DNA heteroduplexes. PCRamplification proceeded with 2 cycles at 95°C for 15 s and60°C for 20 s followed by 38 cycles at 90°C for 15 s and 60°Cfor 20 s and a final extension step of 4 min at 60°C. The

reaction mixtures were then held at 15°C. The total elapsedtime for the RT and subsequent 40-cycle DNA amplificationwas less than 90 min. Omission of the 70°C RT step, whicheffectively prevents cDNA synthesis without compromisingamplification of DNA, was used to distinguish amplificationof DNA from RNA.Upon completion of the amplification reaction, 4 p.l of

each reaction was analyzed by electrophoresis through a1.7% agarose gel in Tris-borate-EDTA buffer (pH 8.0) andethidium bromide staining. DNA was transferred from thegels onto BioTrans nylon membranes (ICN, Irvine, Calif.)by alkaline transfer (20). The transferred DNA was cross-linked by UV light at the autocross-link setting of aStratalinker (Stratagene, San Diego, Calif.). The blots wereprehybridized with a solution containing 5 x SSPE (1x SSPEis 0.18 M NaCl, 10 mM NaPO4, and 1 mM EDTA [pH 7.7]),Sx Denhardt's solution, 100-p.g/ml sheared denatured her-ring sperm DNA, and 0.1% sodium dodecyl sulfate (SDS) at50°C for 1 h and then were hybridized at the same temper-ature for 1 h with probe KY88 labelled with 32P by usingpolynucleotide kinase (Boehringer Mannheim; Indianapolis,Ind.). The blots were washed twice at room temperaturewith 2x SSPE-0.1% SDS and then were washed twice for 15min at 65°C with the same buffer. The blots were autoradio-graphed without an intensifying screen with Kodak XARX-ray film at 25°C for 3 h. Alternatively, amplificationproducts can be hybridized in solution to probes immobi-lized in wells of microtiter plates and detected colorimetri-cally (16).

Clinical samples and sample preparation. Seroconversionpanels from three HCV-infected donors were obtained fromSerologicals, Inc. (Clarkston, Ga.). The panels are com-posed of plasma samples collected serially over a period of 1year. All three donors were infected through the injection ofpacked erythrocytes from the same index case. (The eryth-rocyte injections were administered to generate antibodyagainst Rh factor.) Patient 1 received a single injection.Patient 2 received two injections 25 days apart. Patient 3received three injections; the second and third injectionswere given 1 and 3 days, respectively, after the initialinjection. The volume of cells injected each time was 2 ml orless. Plasma from the index case, collected at the same timeas the erythrocytes, initially tested negative for HCV anti-body by a prototype test but was subsequently determined tobe positive by a Food and Drug Administration-approvedtest (Ortho Diagnostics, Rahway, N.J.). Plasma samplesobtained from the index case on later dates also werepositive for HCV antibody. The index case plasma exhibitedno elevation in serum alanine aminotransferase (ALT) levelat the time of original donation or thereafter. Information onthe size and nature of inocula, antibody status, and levels ofALT in serum was provided by Serologicals, Inc. It shouldbe noted that the plasma samples as supplied by Serologi-cals, Inc., had been stored under less-than-ideal conditions.All specimens were subjected to at least two rounds offreeze-thaw prior to testing, and some specimens werestored at 4°C for extended periods.

Total nucleic acids were isolated from 100-p.l aliquots ofeach serial plasma sample by using the IsoQuick nucleic acidextraction kit (Microprobe Corporation, Bothell, Wash.).Briefly, virus was lysed with a solution containing guanidiumisothiocyanate, and the lysate was incubated at 65°C for 10min. Proteins were then extracted twice with an extractionbuffer and an organic extraction matrix slurry. Nucleic acidswere recovered from the aqueous phase by precipitationwith isopropanol, washed sequentially with 70 and 95%

VOL. 31, 1993

on August 10, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

884 YOUNG ET AL.

RT-PCR PCR

Negative HI0 HCV RNA HCV RNA HCV DNAM Plasma,,,, .2 Ma b c dMe f g hMi j k I M

FIG. 1. Analytical sensitivity of HCV RNA amplification by thecombined RT-PCR assay. A 10-fold dilution series of HCV RNAstranscribed in vitro from the recombinant vector pHCV1.1A wasamplified by combined RT-PCR to assess the analytical sensitivityof the assay. As a comparison, a similar dilution series ofpHCVI.1A DNAs was amplified. Control reactions in which RNAtemplates were amplified in the absence of an RT step were includedto demonstrate that amplification was from RNA templates and notresidual plasmid DNA. Lanes: a through d, amplification of 104, 103,102, and 101 copies of HCV RNA transcripts, respectively, bycombined RT-PCR; e through h, amplification of HCV RNA tran-scripts shown in lanes a through d in reactions in which the RT stepwas omitted; i through 1, amplification of 10', 103, 102, and 101 copiesof pHCV1.1A DNA, respectively; M, 123-bp ladder size markers.The bottom panel is a Southern blot of the gel in the top panel.

ice-cold ethanol, and then air dried. The nucleic acids wereresuspended in 20 ,ul of water treated with diethyl pyrocar-bonate and stored at -70°C. Two microliters of each sample(equivalent to the nucleic acids isolated from 10 ,ul ofplasma) was amplified in the combined RT-PCR assay induplicate. The samples were not blinded in this study.However, at least two separate nucleic acid extractions andRT-PCR analyses were performed for each specimen.Appropriate precautions were taken to avoid sample-to-

sample carryover and contamination by cloned DNA or PCRproducts during both the sample preparation and amplifica-tion setup steps (15). Negative control plasma samples wererandomly interspersed with clinical samples during samplepreparation and carried through amplification. In addition,no-nucleic-acid controls were included with each set ofamplification reactions.

RESULTS

Analytical sensitivity of combined RT-PCR assay. In orderto assess the analytical sensitivity of amplification, a 550-ntfragment from the 5' end of the HCV genome was clonedinto a transcription vector under the control of an SP6promoter. The concentration of HCV RNA transcribed invitro from the plasmid pHCV1.1A was determined, andknown amounts of RNA transcript were used as templates inamplification reactions. Amplification signals from 100 RNAmolecules were seen consistently by gel electrophoresis(Fig. 1). Occasionally, weak signals from 10 RNA copieswere also seen (data not shown). Hybridization to probes

immobilized on microtiter plates allowed the consistentdetection of 10 RNA copies (data not shown). Amplificationof nucleic acids from a serum whose viral titer was deter-mined by infection of chimpanzees (kindly provided by A.Prince) yielded a titer that is 10-fold higher than the titerdetermined previously (data not shown). This level of sen-sitivity is comparable to those achieved by nested PCRsystems reported by others (7).The efficiency of DNA amplification was determined by

parallel amplification with the pHCV1.1A DNA as template.The signal from 10 copies ofDNA was roughly equivalent tothat from 100 copies of RNA (Fig. 1). This suggests thatcDNA copies are generated from only a fraction of the RNAtemplates during the RT step. To ensure that products fromRT-PCR resulted from amplification of RNA rather thanamplification of residual DNA present in the RNA prepara-tion, amplification reactions in which the RT step wasomitted were performed. As shown in Fig. 1, lanes e throughh, no products were generated from as many as 106 RNAcopies when the RT step was omitted. At input RNA levelsof 108 or higher, amplification products were seen in theabsence of RT (data not shown), suggesting the presence ofresidual DNA or a very low level of RT activity during DNAamplification.

Pattern ofRNA detection in HCV-infected patients. Nucleicacids were extracted from plasma samples collected seriallyover a period of 1 year from three individuals infected withHCV. The nucleic acid from the equivalent of 10 p.l ofplasma was amplified by the combined RT-PCR assay.Figure 2 summarizes the results from ALT determinationand antibody and RNA testing for all three individuals. Allthree subjects shared similar patterns of viremia, ALTelevation, and seroconversion. HCV RNA was detectedwithin 10 to 14 days of exposure to HCV. RNA remainedpositive until seroconversion. At about that time, the viralRNA level fell to below the 100-copy limit of detection by gelelectrophoresis and Southern blot hybridization in the 10 ,ulof plasma tested, a titer of < 104/ml. These samples were notanalyzed in microtiter plates. Elevated ALT levels were firstevident 38 to 42 days postinfection. ALT levels returned tonormal 114 to 140 days postinfection and remained normalfor the remainder of the follow-up. Antibody to HCV wasfirst detected 46 to 113 days after exposure, and the samplesremained positive thereafter. Representative results of RNAamplification by combined RT-PCR from patient 1 are shownin Fig. 3. A band of higher molecular weight is consistentlyseen both in these clinical samples and in the cloned tem-plates (Fig. 1). The origin of this band is not known, but theband is HCV-specific, given its ability to hybridize to theHCV probe.

DISCUSSION

We have developed a combined RT-PCR assay in whichHCV RNA can be reverse transcribed into cDNA and thenamplified by PCR under a single set of conditions. This assayis superior to conventional methods for the amplification ofRNA by PCR in that the need to change or add reagentsbetween the RT and PCR steps is eliminated. This obviatesthe need to open reaction tubes after initial setup, minimizesthe possibility of contamination, and reduces the hands-onwork required. The use of a thermoactive enzyme alsopermits RT to be carried out at elevated temperatures, whichincreases the specificity of priming and increases efficiencyby destabilizing secondary structures in the RNA template.In addition, the high enzyme concentration utilized for

J. CLIN. MICROBIOL.

on August 10, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

HEPATITIS C VIRUS RNA DETECTION BY COMBINED RT-PCR 885

a-

M1.1.2a

.2

Patient 1

HCVRNA *eeeooeoeooooooooooooo600500

400300100.

Patient 2

HCVPRNA0000000-oooooo00000000000000

400300200100

.. .....

Patient 3

HCV RNA120011001000300

4003002001000

0 * *0oo0 0 oo ooo oooomcvantjo= ~~~~~~~~~~~~~~~~~I

.. .. .. .

't m C4%a & en Q1; t a fA!R* i

Day post-infectionFIG. 2. Patterns of HCV RNA, serum ALT level, and HCV

antibody in the three study subjects. Solid circles indicate that HCVRNA was detected consistently, half-filled circles indicate thatspecimens were positive for HCV RNA in only some of the replicateanalyses, and open circles indicate that HCV RNA was consistentlynot detected at the limit of 100 copies in the 10 .1l of plasma tested.

efficient RT permits short extension times to be used duringDNA amplification, greatly reducing the total time requiredfor the assay.

Determination of the efficiency of RT-PCR assays asapplied to the detection of HCV RNA has been hampered bythe lack of a method for growing large quantities of virus.Chimpanzee infectious dose titered virus stocks are notuseful for this purpose because only infectious virus ismeasured. While measurement of infectious virus may beclinically relevant, it does not necessarily correlate withvirus particle count, as demonstrated by us and others (9).To obtain a direct measure of the analytical sensitivity of ourcombined RT-PCR assay, we have constructed a transcrip-tion vector from which HCV RNA is transcribed in vitro andthe RNA concentration is determined photometrically byUV A260. By using such well-characterized templates, theanalytical sensitivity of the combined RT-PCR assay wasfound to be 100 copies of RNA templates when analyzedby gel electrophoresis. Hybridization to the probe improvesthe analytical sensitivity to 10 RNA copies. This level ofsensitivity was achieved without a need for the nested

Patient fgmm

Ma b c d e f' 2 h X j k i mn opq r s t u vM

-244 bp

FIG. 3. Representative results of combined RT-PCR amplifica-tion of plasma samples from patient 1. Lanes a through v correspondto days postinfection as in Fig. 2, patient 1. Lanes M, 123-bp laddersize markers. The bottom panel is a Southern blot of the gel in thetop panel.

amplification reported by others, in which two pairs ofprimers were used in sequential amplification reactions (8,18). Nested PCR did not increase the detection limit of thecombined RT-PCR assay (data not shown). The need fornested PCR in most systems reported to date points to theinefficiency of these assays and highlights the need forwell-characterized RNA templates to optimize amplificationconditions. Furthermore, the use of two separate primerpairs in sequential amplification reactions during nested PCRrequires the transfer of reagents during amplification. This islabor intensive and increases the likelihood of contamina-tion.To demonstrate the clinical utility of the combined RT-

PCR, we examined the time course of viremia in threeindividuals. Plasma samples collected serially over a periodof 1 year were examined for the presence of HCV RNA.Similar patterns of transient viremia early in infection fol-lowed by a decline to levels below 100 molecules in 10 ,ul ofplasma at about the time of seroconversion were seen in allthree subjects. This pattern of transient viremia has beenreported in cases of acute, resolving HCV infections inhumans and chimpanzees (7, 8). The fact that all threeindividuals in this study were infected by the same indexcase and showed similar patterns of viremia is consistentwith the hypothesis that the infecting virus is an importantfactor in determining the natural history of HCV infection(7). However, interpretation of the clinical significance ofthese results must be made in light of the small sample size(three subjects) and less-than-ideal storage conditions of thespecimens (3). Curiously, HCV RNA was not detected in thesource plasma from the index case (data not shown). Thissuggests that the concentration of virus may be below 100RNA copies in the 10 ptl of plasma tested (the equivalent of104 copies per ml). This implies that the infectious dose ofHCV may be low, because each subject was exposed to nomore than 2 ml of packed erythrocytes. Subsequent samplesfrom the index case taken at 3, 5, and 7 months after theinitial donation were positive for HCV RNA.

VOL. 31, 1993

on August 10, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

886 YOUNG ET AL.

The combined RT-PCR assay described in this report is auseful tool for the detection of HCV RNA. The exquisitesensitivity of the assay is ideal for monitoring patientsundergoing antiviral therapy to determine the treatment endpoint. With the addition of appropriate internal controls, theassay also has the potential for quantitative analysis of virusload, useful in determining treatment regimen and patientmonitoring. In addition, the assay can be applied to theamplification of any RNA template. These include otherRNA viruses (such as human immunodeficiency virus) andmRNA. In situations involving detection of retroviral RNAand mRNA, care must be taken to ensure that amplificationis from RNA templates. In an assay with the sensitivity ofRT-PCR, amplification of residual DNA can lead to prob-lems in the interpretation of results, especially where largevolumes of specimen are tested.

ACKNOWLEDGMENTS

We thank the RMS DNA synthesis group for synthesis of theoligonucleotides. The HCV-infected serum used in the initial cloningof HCV sequences was kindly provided by H. Fields of the CDC.We also thank J. Sninsky and D. Gelfand for critical review of themanuscript and helpful discussions.

REFERENCES1. Alter, H. J., R. H. Purcell, J. W. Shih, J. C. Melpolder, M.

Houghton, Q.-L. Choo, and G. Kuo. 1989. Detection of antibodyto hepatitis C virus in prospectively followed transfusion recip-ients with acute and chronic non-A, non-B hepatitis. N. Engl. J.Med. 321:1494-1500.

2. Buell, G. N., M. P. Wickens, F. Payvar, and R. T. Schimke.1978. Synthesis of full length cDNAs from four partially purifiedoviduct mRNAs. J. Biol. Chem. 253:2471-2482.

3. Busch, M. P., J. C. Wilber, P. Johnson, L. Tobler, and C. S.Evans. 1992. Impact of specimen handling and storage ondetection of hepatitis C virus RNA. Transfusion 32:420-425.

4. Chamot, E., B. Hirschel, J. Wintsch, C.-F. Robert, V. Gabriel,J.-J. Deglon, S. Yerly, and L. Perrin. 1990. Loss of antibodyagainst hepatitis C virus in HIV-seropositive intravenous drugusers. AIDS 4:1275-1277.

5. Choo, Q.-L., G. Kuo, A. J. Weiner, L. R. Overby, D. W.Bradley, M. Houghton. 1989. Isolation of a cDNA clone derivedfrom a blood-borne non-A, non-B viral hepatitis genome. Sci-ence 244:359-362.

6. Cristiano, K., A. M. Di Bisceglie, J. H. Hoofhagle, and S. M.Feinstone. 1991. Hepatitis C viral RNA in serum of patients withchronic non-A, non-B hepatitis: detection by the polymerasechain reaction using multiple primer sets. Hepatology 14:51-55.

7. Farci, P., H. J. Alter, D. Wong, R. H. Miller, J. W. Shih, B. Jett,and R. H. Purcell. 1991. A long-term study of hepatitis C virusreplication in non-A, non-B hepatitis. N. Engl. J. Med. 325:98-104.

8. Garson, J. A., R. S. Tedder, and M. Briggs. 1990. Detection ofhepatitis C viral sequences in blood donations by "nested"polymerase chain reaction and predicted infectivity. Lancet335:1419-1422.

9. Garson, J. A., P. W. Tuke, M. Makris, M. Briggs, S. J. Machin,

F. E. Preston, and R. S. Tedder. 1990. Demonstration ofviraemia patterns in haemophiliacs treated with hepatitis-C-virus-contaminated factor VIII concentrates. Lancet 336:1022-1025.

10. Godec, M. S., D. M. Asher, P. T. Swoveland, Z. A. Eldadah,S. M. Feinstone, L. G. Goldfarb, C. J. Gibbs, Jr., and D. C.Gajdusek 1990. Detection of measles virus genomic sequencesin SSPE brain tissues by the polymerase chain reaction. J. Med.Virol. 30:237-244.

11. Han, J. H., V. Shyamala, K. H. Richman, M. J. Brauer, B.Irvine, M. S. Urdea, P. Tekamp-Olson, G. Kuo, Q.-L. Choo, andM. Houghton. 1991. Characterization of the terminal regions ofhepatitis C viral RNA: identification of conserved sequences inthe 5' untranslated region and poly(A) tails at the 3' end. Proc.Natl. Acad. Sci. USA 88:1711-1715.

12. Kato, N., M. Hjikata, Y. Ootsuyama, M. Nakagawa, S. Ohko-shi, T. Sugimura, and K. Shimotohno. 1990. Molecular cloningof the human hepatitis C virus genome from Japanese patientswith non-A, non-B hepatitis. Proc. Natl. Acad. Sci. USA87:9524-9528.

13. Kedzierski, W., and J. C. Porter. 1991. A novel non-enzymaticprocedure for removing DNA template from RNA transcriptionmixtures. BioTechniques 10:210-214.

14. Kotewicz, M. L., C. M. Sampson, J. M. D'Alessio, and G. F.Gerard. 1988. Isolation of cloned Moloney murine leukemiavirus reverse transcriptase lacking ribonuclease H activity.Nucleic Acids Res. 16:265-277.

15. Kwok, S., and R. Higuchi. 1989. Avoiding false positives withPCR. Nature (London) 339:237-238.

16. Loeffelholz, M. J., C. A. Lewinski, S. R. Silver, A. P. Purohit,S. A. Herman, D. A. Buonagurio, and E. A. Dragon. 1992.Detection of Chlamydia trachomatis in endocervical specimensby polymerase chain reaction. J. Clin. Microbiol. 30:2847-2851.

17. Myers, T. W., and D. H. Gelfand. 1991. Reverse transcriptionand DNA amplification by a Thermus thermophilus DNA poly-merase. Biochemistry 30:7661-7666.

18. Okamoto, H., S. Okada, Y. Sugiyama, T. Tanaka, Y. Sugai, Y.Akahane, A. Machida, S. Mishiro, H. Yoshizawa, Y. Miyakawa,and M. Mayumi. 1990. Detection of hepatitis C virus RNA by atwo-staged polymerase chain reaction with two pairs of primersdeduced from the 5'-noncoding region. Jpn. J. Exp. Med.60:215-222.

19. Okamoto, H., S. Okada, Y. Sugiyama, S. Yotsumoto, T. Tanaka,H. Yoshizawa, F. Tsuda, Y. Miyakawa, and M. Mayumi. 1990.The 5'-terminal sequence of the hepatitis C virus genome. Jpn.J. Exp. Med. 60:167-177.

20. Reed, K. C., and D. A. Mann. 1985. Rapid transfer ofDNA fromagarose gels to nylon membranes. Nucleic Acids Res. 13:7207-7221.

21. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecularcloning: a laboratory manual, 2nd ed., p. 7.26-7.29. Cold SpringHarbor Laboratory Press, Cold Spring Harbor, N.Y.

22. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecularcloning: a laboratory manual, 2nd ed., p. E.5. Cold SpringHarbor Laboratory Press, Cold Spring Harbor, N.Y.

23. Weiner, A. J., G. Kuo, D. W. Bradley, F. Bonini, G. Saracco, C.Lee, J. Rosenblatt, Q.-L. Choo, and M. Houghton. 1990. Detec-tion of hepatitis C viral sequences in non-A, non-B hepatitis.Lancet 335:1-3.

J. CLIN. MICROBIOL.

on August 10, 2020 by guest

http://jcm.asm

.org/D

ownloaded from